Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    unrelated or irrelevant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    有一些
    0.51
     همیشه
    0.45
     adhere
    0.43
    ยังมี
    0.42
     Jupyter
    0.41
     Compliance
    0.40
     comply
    0.40
     complied
    0.40
     compliance
    0.40
    以前
    0.39
    POSITIVE LOGITS
     unrelated
    0.91
     irrelevant
    0.76
    无关
    0.65
    UNRELATED
    0.52
     전혀
    0.48
    何の
    0.48
     никакого
    0.47
     ámbito
    0.46
     relevance
    0.46
     focuss
    0.46
    Act Density 0.203%

    No Known Activations